Onyx Security said on March 12 it raised $40 million in seed and Series A funding led by Conviction and Cyberstarts as it launches a security control plane designed to govern enterprise AI agents. The company’s financing totals $40 million, including a $35 million Series A, according to Israeli business outlets Globes and Calcalistech.
The platform targets the growing risk surface of autonomous agents by continuously monitoring behavior and enforcing policy. Onyx says its Guardian Agent can detect abnormal activity and respond by blocking actions, reducing permissions, or requiring human approval before a task executes.
Onyx was founded in 2024 by Maxim Bar Kogan and Gil Elbaz and now has more than 70 employees across Israel, the U.S., and Canada, according to the two reports. The startup is positioning itself as infrastructure for enterprise AI adoption rather than a point security product.
In its launch announcement, Onyx framed agent adoption as a new operational risk as reasoning errors and prompt-based attacks can trigger unintended actions. The $40M raise suggests investors see a market forming around “agent control planes” that give security and governance teams visibility, policy enforcement, and runtime guardrails as organizations scale automation.
Key points
- Onyx Security raised $40M led by Conviction and Cyberstarts to launch an AI-agent security control plane.
- The Guardian Agent can detect abnormal activity and respond by blocking actions, reducing permissions, or requiring human approval.
- The company positions itself as infrastructure for enterprise AI adoption, with 70+ employees across Israel, the U.S., and Canada.
AI Signals coverage: https://1m-reviews.com/category/ai-signals/
